The pharmaceutical industry is facing a tremendous challenge in the drug development and testing pipeline; many questions and concerns have been raised about the efficiency effectiveness of current drug screening approaches. The cost of bringing a single drug to market is now estimated at a billion dollars and a major portion of this money is invested in preclinical stages where researchers test for the efficacy and safety of drugs. The current gold standards in such testing are a) animal testing b) simple in-vitro models of different organs where liver is considered the most important target both in terms of its drug metabolic functions and susceptibility to toxicity. Unfortunately, these current gold standards have low predictive capabilities which necessitated current research initiatives both in US and EU to replace those with microfluidic tissue models; these are expected to provide better physiological recapitulation and thereby better toxicological predictivity of new compounds. Our group among a few others are developing such models. Nevertheless, no current models take into account the zonation in the liver - in a controlled and actively enforced fashion - which can give rise to drastic differences in drug metabolic and detoxifying activity across the liver as wel zone specific toxicity. Accordingly in this application we aim to address this gap and build on our liver tissue engineering and microfabrication expertise towards developing a novel microfluidic platform that incorporates an actively controlled liver zonation. We will achieve this by integrating several key advances developed in our lab such as a multi-layer microfluidic culture device, actively controlled gradient generator and an ultra-thin collagen coating. In aim 1, we wil first use this newly developed platform to characterize the zonation in a mono-culture of human hepatocytes and then demonstrate the zone-specific response of this platform by challenging it with drugs known to induce zone-specific toxicity. In aim 2, we will expand this platform to include human liver sinusoidal endothelial cells and will then gauge the effect of interaction between hepatocytes and the endothelial cells on both cell types when the system is challenged in a similar fashion as in aim 1. The work, we describe here, is expected to establish the foundation of a realistic microfluidic liver co-culture model which controllably recapitulates zona response within a liver sinusoid. This is a significant advance that will not only enable better prediction of drug responses but will also provide high fidelity fundamental information about liver physiology. This will result in reduction of drug development costs as well as number of drugs that make it to the market with harmful effects; both results will greatly be beneficial to te general public.